Every enterprise has access to AI now, so access isn't the differentiator anymore. The harder problem is turning a model that can reason into work that actually gets done, work that stays inside the rules and holds up across systems that were never designed for autonomous software. That takes two capabilities most organizations are still buying as if they were unrelated. The first is an AI platform to build intelligence into real applications and agents. The second is an orchestration layer to harness that intelligence and carry it through to governed, completed work.
AI powers transformation, and orchestration harnesses it. Both are critical for operations.
The AI half: A platform, not a model
AI models are becoming commoditized and the true driver of value in the AI world will be the platforms which enable you to confidently scale LLM's far and wide. The Forrester Wave™ report puts it plainly when it says “The previous decade of enterprise AI centered on automation through insights and alerting. But agentic applications create value by understanding and acting across complex workflows and environments.”¹
So the AI half of the equation isn't just whether you're running models, or which you’re using. It's whether you have a platform that can build AI into your business and govern it once it's running. That's a distinct capability. Not the raw intelligence, but the platform that makes the intelligence usable.
The orchestration half: Workflows as the harness
The second half tends to get underestimated. Even a strong AI platform produces agents that reason and act, and any agent that acts needs a harness: the scaffolding around the model that decides what it's allowed to do, feeds it the right context, calls the right tools, and keeps its behavior inside bounds. Left on its own, a model just produces output. The harness is what turns that output into an action the business can rely on.
The concept is new enough that the tooling is still settling. Open-source frameworks like LangChain and LangGraph arrived early and became a common way to wire an agent's reasoning loop to its tools and memory, and they're a reasonable place to prototype. What enterprises keep discovering as they move past the prototype is that a lightweight harness doesn't hold up under real conditions. Once an agent is touching regulated data or customer money, the harness has to carry persistent state across work that runs for weeks, produce an audit trail, enforce security and access policy, and apply the business rules every time rather than most of the time. That's a much higher bar than an agent framework was built to clear.
That higher bar already has a name: enterprise orchestration. It enforces business rules deterministically, holds the state of a case that can run for months, knows when a human has to step in, and produces the audit trail a regulator will eventually ask for. Workflows are how intelligence gets pointed at a real outcome and kept inside the guardrails on the way there. This is what the business orchestration and automation technology (BOAT) category exists to provide, and it's the difference between an agent that can run in production and one that only runs in a demo.
Why you need both
Put simply, an AI platform supplies the intelligence, and orchestration supplies the control that makes the intelligence safe to use at scale. Take either one away and the other underdelivers. An AI platform without orchestration produces impressive output that nobody can trust inside a regulated process, while orchestration without a real AI platform produces well-governed workflows that can't reason or adapt. The value only shows up when the two reinforce each other.
The catch is that they've historically come from different vendors built on different foundations. AI-first platforms tend to be strong on reasoning and light on durable, governed process, while process platforms tend to be strong on control and, until recently, light on native AI. Buy one of each and wire them together, and you create a seam between them. That seam is where regulated, mission-critical work tends to break, because a hand-off loses its audit trail or a decision runs without a rule to vet it. It's also where the return on AI tends to disappear. Enterprises have put tens of billions into generative AI and most executives still report no financial return so far,² and those failures are rarely about the model itself. They happen where the two systems meet.
Which is why the position worth taking is the harder one. You need both, and you need them working as a single system so there's no seam in the middle. Being genuinely strong in both is rare, because it means competing well on two different fields that most vendors only play on one of: the AI platform, and the orchestration that harnesses it.
Pega is recognized as a Leader in both.
Pega Named a Leader in The Forrester Wave™: AI Platforms, Q3 2026. Pega received the highest score possible in 8 of 19 evaluation criteria. Read the report →
Pega Named a Leader in the 2026 Gartner® Magic Quadrant™ for BOAT Platforms. Pega recognized as a Leader for the second year in a row. Read the report →
Forrester does not endorse any company, product, brand, or service included in its research publications and does not advise any person to select the products or services of any company or brand based on the ratings included in such publications. Information is based on the best available resources. Opinions reflect judgment at the time and are subject to change. This report is part of a broader collection of Forrester resources, including interactive models, frameworks, tools, data, and access to analyst guidance. For more information, read about Forrester’s objectivity here.
Sources
- Forrester Research, "The Forrester Wave™: AI Platforms, Q3 2026," Mike Gualtieri and Rowan Curran with three contributors (Sudha Maheshwari, Kylie Cadogan, and Jen Barton). Evaluated 15 providers; vendor materials cutoff May 19, 2026.
- PwC, "29th Annual Global CEO Survey," January 2026, https://www.pwc.com/gx/en/news-room/press-releases/2026/pwc-2026-global-ceo-survey.html; investment scale per MIT Project NANDA, "The GenAI Divide: State of AI in Business 2025," July 2025, https://mlq.ai/media/quarterly_decks/v0.1_State_of_AI_in_Business_2025_Report.pdf.
- Pegasystems, press release, "Pega Named a Leader in AI Platforms by Independent Research Firm," August 6, 2026, https://www.pega.com/about/news/press-releases/pega-named-leader-ai-platforms-independent-research-firm (source for the "highest score possible in 8 of 19 criteria" figure in the recognition lockup).